103 results in /b/ - Boson Technology Anonymous 09/02/24 (Mon) 02:52:53 No. 12848
>>12840 I kind of like the gimmick of rotating boards, but I get some people might get frustrated if a conversation gets frozen midway through. I don't really care what it's called as long as the role of 3D/random is filled somewhere.
Most threads on /secret/ I'd prefer to see on /jp/ or /qa/, including many of the "controversial" topics. What's left is probably youtube spam and blogging, maybe some autistic ranting, which is all fine deleted or hidden away in my eyes. Posts on real boards either are okay or they're not. Half-measures just piss people off when a basically okay thread gets pseudo-deleted and encourages actually kind of good threads to be started hidden away there just in case they might trigger the mods.
/maho/ doesn't really hurt anything if you use /all/, but I also don't think it will justify itself at the current rate. There's not a community of highly skilled people focused on tech discussion here to justify needing their own space, nor is there enough serious generalist tech talk to draw the post numbers that justify a split. Maybe it could get there with a few dedicated posters, but I just don't see it happening.
Anonymous 06/25/24 (Tue) 06:25:29 No. 12734
>>12730 Isn't the fact that people are less likely to look at older replies a problem in of itself? Though, for long lasting threads (which this site has a lot of), I suppose focusing on recent replies is good to foster activity.
Anonymous 06/13/24 (Thu) 02:11:09 No. 12639
we're thinking about names for kissu's technology focused board. I'm thinking of /98/ or /pc98/ to keep it in line with japanese interests and give it some crossover with /qa/ and /jp/. Other opinions voiced have been /l/ which I fear is going to wind up looking very odd when people start asking about tor questions and be unhinged about their fetishes. It does have crossover with ideas like lain and linux.
Anonymous 06/08/24 (Sat) 05:49:49 No. 12611
>>12608 >>12610 I understand what you're saying as far as moving
current threads goes, but I do not see how axing the seasonal boards achieves in making /qa/ more active and diverse in topics going forwards (As
>>12509 and
>>12514 described).
Maybe you disagree with my diagram showing board overlap (
>>12607 ), but as I see it, /secret/ is currently siphoning off posts from the seasonal boards, which themselves were originally siphoning off-topic posts from /qa/ and /jp/ pre-/secret/, post-split, most notably in the pseudo blogs ("free thoughts", "cafe", etc.). If the seasonal boards are archived, I really think that /secret/ should be unlisted and maybe its name changed to something more representative of what should go there, because I really think a lot of the activity the seasonal boards have would instantly migrate to /secret/; quite frankly, I've somewhat come around to /secret/. At this point, /qa/ really does not resemble 4/qa/ at all, whereas /secret/ is increasingly become more /qa/-like as posts that would have been made elsewhere feel unwelcome; Making a thread about anything all at "unfocused" feels clumbersome and misplaced on /qa/.
Kissu's /qa/ is now more of a "Project/General" board, and /secret/ is increasingly what 4/qa/ was, strife included... If I had any power at all, I would rename /qa/ to /pg/ - Project/General, list /secret/ and then rename to /qa/. At that point, separating technology threads from /qa/ makes more sense on an organizational level, even if I do think the board will be very low activity.
Anonymous 06/07/24 (Fri) 13:57:40 No. 12555
>>12554 >Demand for something that exists As far as I can tell, this only cropped up just now, and there's not particularly any strong opinions in this thread other than discussing what it would look like if it does indeed become a board. /b/ used to be the tech board until someone wanted to nuke it all and make it into their personal blogging and meta board, but even then the number of threads wasn't particularly high compared to the number of people misinterpreting the /b/ name as being the random board. The existence of a linux thread, a PC thread, and other tech-related threads is not proof of desire for a tech-focused board. I also tend to agree with
>>12534 and
>>12535 .
>a generalification of /qa/ as a result of the seasons board The reason /qa/ is filled with mega threads is not because of the seasonal boards. This should be obvious. /qa/ is filled with mega threads because of the absence of /jp/-like threads that contributed to board churn that made it feel more active and spontaneous. Archiving the seasonal boards isn't going to put those thread ideas back into /qa/. It's either going to result in more consolidation into the blog thread and similar large threads, or those off-topic posts are going to show up on /jp/ or /secret/ because they're perceived as having a lower barrier to entry.
Further removing technology threads is again just going to entrench /qa/ as a general thread board until there's no topics left to discuss that can't be had elsewhere. You cannot go back to pre-/jp/ split /qa/ when the differentiation between /qa/ and /jp/ (and others) is exactly what you are looking at! /qa/ is the way it is because the "/qa/" you want to put back into it is on /jp/ and elsewhere. The seasonal boards are not the cause. This is particularly the case after it was decided to nuke any blog-like threads from the seasonal boards.
This is not to say that removing the seasonal boards would do nothing. It will. It just won't change anything on /qa/ in particular.
>desire to move things away from IRC Then remove /chat/...
Anonymous 06/06/24 (Thu) 20:18:10 No. 12517
Thought the solstice was in July so I'd have to do something before then.
Anonymous 03/06/23 (Mon) 15:55:54 No. 11540
no no interpretation is relatively easy
it's that it's a hodpodge of weak ideas thoughtlessly strewn about with pretensions of originality
for example
>In my opinion the strength of an online community is not determined by the strength of software in the above. default stance across most of the internet, borderline redundant
>IM’s spontaneity can’t inspire hundreds of posts in an hour between two engaged individuals. wrong, especially in its timeframe
>Indeed, MBs can produce their own Activities and feedback loops. verbose but shallow, that's called a conversation
>However, a set of users focusing on the MB or talking heads in an IM server at typically creates meta entertainment which tends to create ingroups and communities which inherently gatekeep newcomers from taking part. "cliques r bad", idiosyncratic concept of meta whose meaning isn't delineated anywhere
>A community which does not focus on Activities does not exist. tautological
>They may discuss things outside of the game, but ultimately everything interesting comes back to the game itself. wrong again, off-topic posts may be interesting yet disconnected from the main topic despite taking place in a platform dedicated to it, arguing otherwise would require an extremely liberal conception of "coming back to"
Post too long. Click here to view the full text. Anonymous 01/25/23 (Wed) 19:08:02 No. 11322
What you have to remember is that Kissu isn't just 2D; it's 2D/random , being an extension of that culture on 4/qa/. The 2D side of things provides the site with a core focus that everything loops back around to, but we've never been afraid to stray from that focus for the sake of an interesting discussion.
Anonymous 01/25/23 (Wed) 02:31:44 No. 11309
Thinking about your thread a bit. I don't think there's an issue with the type of content on Kissu. It's mostly an issue that the otaku medium is sort of in a weak place right now.
Anonymous 01/24/23 (Tue) 06:09:56 No. 11288
>>11285 I'm far too old for these purity tests.
Good people liked watching trash TV and it hasn't changed after streaming replaced network cable. My dad was one of the smartest people I've ever known and he watched the absolute most god-awful horrendous trash that you'd assume anyone would need brain damage to watch. This is a guy that would work on a B-52 or the freakin' space shuttle and then come home to watch Jerry Springer.
You can focus on the stuff that bothers you, but it's going to be a terrible experience. It's how the rest of the internet works, after all. If you think that thread has bad videos, then post good ones to make it better. I generally don't post much in there myself since I'm more of a news reader than watcher and it's rare to see good documentaries and stuff on youtube.
Seasonal boards are also supposed to involve real life things, like the seasons, and they can be easily hidden from view forever. This became a thing because it was bothering at least one person that a major event of late 2019/2020 was ruining the mood on /qa/, so instead we ask people to post that stuff on seasonal boards.
Kissu has layers of escapism built into it and it's your choice as to how far you will go.
Anonymous 01/24/23 (Tue) 05:44:52 No. 11285
>>11282 >>11283 >>11284 I'm using /secret/ as an example of such contents, but it's not the only place they exist. I can also point to the "random video" threads on seasonal boards where lots of video thumbnails prominently contain human reactions.
The fact that they posted these videos rather than the alternatives that don't involve reactions, which are also posted in the thread (and thus focusing on presenting facts) shows that these posters are easily swayed by the reaction of 3D internet celebrities. Not to mention most of these videos are about real-life topics in the first place.
Anonymous 11/04/22 (Fri) 20:12:35 No. 10957
In a week or so I'll be switching my dev /b/log format into the following format of multiple generals style threads. Likely they will all be stickies:
- Vichan rewrite Tsukuyomi (TBA)
- UI Projects Kissu-UI and Lite-UI (TBA)
- Financial talk (
>>6583 )
- Feedback&Forms Project (
>>10557 )
- Banners Project (
>>7514 )
- Rules, FAQ things... Managerial talk (
>>10758 probably a new thread though)
Biggest thing will be that:
- Discussion on projects will be sorted into these categories and misplaced things will go into other locations.
- More stickies
- Each one will have a feedback form with a good OP which I'll check minimum once a week.
- Likely no new dev blog threads. If there's a problem they'll be handled like the happening thread.
This change reflects a more mature age of the site where software is established and projects are the focus rather than work on the UI or Server. Mangement projects, dev projects, financial projects etc etc etc
Anonymous 11/04/22 (Fri) 20:06:23 No. 10956
- After testing how Twig templates(the Vichan and PHP style of templating) function in Golang I'm going to be rewritting templates from that standard into Pongo2.
Anonymous 10/13/22 (Thu) 10:02:15 No. 10883
Finally finished a working version and have it running on the site.
Going to use it in something soon.
Still want to upgrade it but I ought to switch focus to some other items for the time being
source code:
https://github.com/ECHibiki/Kissu-Feedback-and-For Anonymous 09/08/22 (Thu) 05:57:29 No. 10751
>>10748 >> The QR box is not in focus when opening with the 'Q' hotkey (This is not limited to the 'Q' hotkey, it seems that the QR box is never in focus when opened anew.). Also, despite probably being redundant, I'd like to add that the "New Thread" box also has the same problem.
Anonymous 09/08/22 (Thu) 05:57:12 No. 10750
>>10749 probably. The way vichan set it up is that it would only do it on jpeg. I think I hardcoded in the same with .png, but probably wouldn't for jfif. It's only there for twitter images and Twitter strips everything.
>>10747 Found it in this terrible search feature which I plan to replace next month...
https://kissu.moe/search.php?search https://kissu.moe/b/res/8005#8804 Anonymous 09/08/22 (Thu) 05:52:53 No. 10748
>>10747 might have thought "it doesn't work on vichan and I need to do something else so I'll ignore it until someone brings it up again"
It's the next item on my list:
> The QR box is not in focus when opening with the 'Q' hotkey (This is not limited to the 'Q' hotkey, it seems that the QR box is never in focus when opened anew.). > Spellcheck correction not detected by undo-redo Anonymous 09/08/22 (Thu) 05:44:13 No. 10745
>>10698 This is not limited to the 'Q' hotkey, it seems that the QR box is never in focus when opened anew.
Anonymous 09/06/22 (Tue) 04:49:29 No. 10698
The QR box is not in focus when opening with the 'Q' hotkey.
Anonymous 08/23/22 (Tue) 20:00:01 No. 10619
>>10596 (another staff member here)
Just so people know, vern has a flair for the dramatic and "kissu is getting some detrimental posts affecting the atmosphere negatively" might come out as something like "kissu's identity as a free enterprise of the philosophical marketplace is in grave peril, we must mobilize the benefactors to quell the chicanery afoot". I don't mean this in a bad way, it's just people might read a post like this and think things are in danger of collapse or severe change and get the wrong idea. He's thinking out loud to get some feedback.
I agree that people are getting more agitated and angry at things instead of simply avoiding them, but my stance has been to address it as a user and not a moderator. However, If people start to give bad posts attention instead of good posts, which is happening in a vtuber thread on /jp/, then this does invoke moderator action to purge people from destroying threads because they don't like the subject. There's an entire world (and many fantasy worlds) to talk about, so people should spend their time talking about stuff they like. You'll find that posting earnestly in threads you enjoy will make you happier and enrich the site, while also moving something you hate down on the page. I never, ever want kissu to be a place where you can expect people to enter a thread just to say something like "kill yourself [current popular buzzword]". While I've been typing this someone else took action, so that's good.
Also, I think this thread should remain focused on software and not community stuff.
Anonymous 08/23/22 (Tue) 13:50:00 No. 10600
>>10598 I'll post the mod-log relating to the topic. If you disagree with my mindset then sure
[7:19 AM]
Verniy | Kissu.moe:
If kissu ever wants to talk about something more serious than what it does already, then over half the posters on the site will need to be moderated or else it will look like 4chan or some other non-extremisy imageboard
I mean, we've got a few people who would explode at any serious vtubter threada guy who has exploded against NTR
and instead of try to remove these problems we;ve attempted to integrate with them because you guys refuse to hunt down posters out of some sort of questionable emotional state
So, sad to say, but kissu will always be a site for the low grade content that has no larger artistic or academic value
and it's not like they need to be banned, but they need to be told that they're being disruptive and the only person doing that is me. You guys just handhold unstable posters and let them continue to be themselves without confronting them
though I would figure it's easier to just remove them and focus on people who matter...
How many land mines need to exist on Kissu before you guys understand the threat?
[7:46 AM]
Verniy | Kissu.moe:
Post too long. Click here to view the full text. Anonymous 08/01/22 (Mon) 05:37:00 No. 10507
>>10506 >go into the name field and press TAB Ah, you were referring to the focus changing from one input field to another.
>This is normal behavior within firefox, and perhaps chrome. Oh.
Anonymous 07/31/22 (Sun) 11:13:21 No. 10495
Ver 4.23.11
Some Dry Kissu Software History Anonymous 06/14/22 (Tue) 10:36:32 No. 10049 [Reply]
A timeline for Kissu's software. Most of this information has been blogged, though the earlier stuff was lost by setting the config options on /b/ to 1 page. December 2018 - Kissu is started out of a need to keep /qa/ going with mod autismo moderation. It picks vichan because that's what all the other spinoffs are using. Verniy gets the site running and picks the cheapest VPS available.
January 2019 - Kissu is released and Verniy works on removing bugs from the software. Begins doing simple changes such as audio hover expansion and so on. On release Captchouli was added to the software in a semi-functional state. Took features from NPFChan for deleting posts to restore bumps. To replace Cytube, theatre.kissu.moe was added to this site.
February 2019 - Kissu has every page checking if posters have been banned. Feature is later removed.
March 2019 - Bug fixes and improvements to the UI. Vichan's catalog post box was probably created at this time.
April 2019 - Kissu releases /agg/. Worked into an April fools joke it was a board that scraped all the /jp/ imageboards with JSON APIs. Stayed around for a few months until it started being difficult.
Around this point in time Kissu took the index updater concept from GNFOS.
May - August 2019 - Not a lot is recorded during this time. Ideas were played with like bans.kissu.moe which was a bullitin board thing. Some concepts of a potential imageboard overboard using /agg/. Had summer courses and a part time job taking up most of my time.
Post too long. Click here to view the full text.
Service Interupts: June 1st V ## Admin 05/31/22 (Tue) 02:36:15 No. 9889 [Reply]
On June 1st posting will have some interrupts while we transfer site information from one location to another. As apposed to the site reformat that happened earlier this year(
>>9095 ), there should not be a half a day long downtime.
Since Kissu is in the last 6 months of it's 4 year server lease our hosts are changing and our server hardware is being upgraded.
Switching from a VPS provider to a Cloud provider.
3X the cost with 3.2X the storage, 2X the RAM and a better CPU (though our former host never gave any details on what they use). Formerly our servers were chosen for being the cheapest option. Currently the focus is on increasing the site's performance and fine tuning our hardware to what makes posting the fastest.
The most important thing of note is that the increase in storage space will resolve an issue with post times that I was working around. Uploading files directly to luna will cease to be a thing(at least until I rework the algorithm).
In addition to the above, but not effecting your posting quality in any way:
A separate VPS server will be purchased and will be uploaded with full backups every 24hr. This ensures that suspensions from server hosts or hardware failure will only ever roll us back a day.
Services such as 4taba, theatre.kissu, file-url-upload and luna.kissu will continue to exist in the same manner as they have been.
Anonymous 04/08/22 (Fri) 20:39:11 No. 9545
design 3 Let m index the pairs of real number bounds, and let n index the pairs of rational number bounds that define each real number. At a particular m and n, you have four rational numbers. There are two real numbers, an upper bound and a lower bound, and at stage n each of those real numbers has a rational upper and lower bound.
The motivation: Notice that if we fix m and increase n, we get increasingly better bounds, but only up to a point. Unless the two real numbers are actually equal, eventually the lower bound on the upper bound becomes larger than the upper bound on the lower bound. This creates a limit on how close the upper bound on the upper bound and the lower bound on the lower bound can get to each other. If the next pair of real numbers (real number pair m+1) has upper bounds on the upper bound and lower bounds on the lower bound that are inside this gap, then the real number m is guaranteed to be no longer useful for computing better approximations to our target value. I've drawn a crude picture to try to explain what I mean.
The idea of design 3 was to increment n at each step, but only increment m if you encounter a situation like the one I just described. And at each such step, use the upper bound on the upper bound and the lower bound on the lower bound as rational bounds on the target value.
But I hadn't entirely thought this idea through, and after further thought I realized that the situation triggering an advance in m wasn't guaranteed to happen, even if real number pair m exhibited a gap preventing further progress. After some further thinking, I came to design 4, which I think will work.
design 4 This is pretty similar to the previous except with a different condition for triggering m to increment. This time, the trigger is when for real number pair m the gap between the lower bound on the upper bound and the upper bound on the lower bound is at least half the distance between the upper bound on the upper bound and lower bound on the lower bound. As before, we use the upper bound on the upper bound and the lower bound on the lower bound as bounds on the target value. A complication is that these bounds are not guaranteed to be nested, but that can be fixed by taking the minimum of all the upper bounds found so far, and the maximum of all the lower bounds found so far.
I think that since this version of the real numbers
Post too long. Click here to view the full text. Anonymous 12/24/21 (Fri) 09:29:21 No. 8907
I think I'm going to start a rewrite along the lines of what I was talking about in
>>7038 . I think I know how to avoid the problems I was having. I've also realized that I can avoid the exponential blowup problem I was talking about by just keeping things simple and just adding/subtracting/multiplying/dividing the sequences termwise. Downside is that some numbers will be calculated to much greater precision than they need to be, but it's not an exponential waste. Maybe at some point in the future I'll implement an efficiency-focused model which represents numbers in some specific base.
The current model is kind of cute despite being impractical for calculations due to the exponential blowup issue, so I'll keep it around on another branch.
I'm also going to try writing out more of my proofs as I go along. I'd like it to be at a level where people can understand more or less what's going on even if they can't read the code. Hopefully this time around there will be more math to write about and less fiddling with technical issues. I'll start with an overview of what the model is and why you might want to model real numbers this way.
Anonymous 12/17/21 (Fri) 21:30:43 No. 8892
>>8885 >in a direction that could focus on political theory Do you have any clear guideline of this? Cause any discussion of history or any act of person, even in a pure theoretical context, could "lead to that direction", even though no one apart from you think so.
Might as well just plainly say "Discussing any real or hypothetical person other then yourself is banned".
Anonymous 12/17/21 (Fri) 21:20:05 No. 8886
>>8885 >was political in nature or in a direction that could focus on political theory. Ponder this: your view that the thread
could become X, can implicitly bias it into becoming as much. Pre-emptive moderation as such may in practice then lead to what you are trying to avoid.
Anonymous 12/17/21 (Fri) 21:15:39 No. 8885
was political in nature or in a direction that could focus on political theory. One of /qa/'s topics is internet culture. Discussion about social/political organizations doesn't fall into that category.
Anonymous 11/16/21 (Tue) 11:30:06 No. 8821
fixed the focus issue
Anonymous 11/13/21 (Sat) 05:01:47 No. 8804
>>8800 I'm not saying it didn't bring up the QR box but that the QR box that appears after clicking on the "Reply" button on the side panel is not automatically in focus. "Automatically in focus" meaning user has to manually click on the QR box that appeared to start typing anything.
Anonymous 11/13/21 (Sat) 03:50:42 No. 8797
>>8796 I've just never used the "Reply" button so I noticed when I used it for the first time. I always click on a post number to get the quick reply box. But I just realized that I had gotten so used to it that I didn't even notice I still have to manually focus the quick reply box.
Anonymous 11/13/21 (Sat) 01:01:58 No. 8795
Something rather minor but after clicking the "Reply" button on a thread (new UI), the popped up quick reply box is not automatically in focus.
Anonymous 10/17/21 (Sun) 01:25:35 No. 8597
Deletions happen now and then to keep threads on track. It's not done against individuals but posts. Anyway, moderation talk is better done on /b/ so threads don't get derailed.
Anonymous 09/08/21 (Wed) 09:20:23 No. 8489
I'll be renamimng , and tech on that board will be moved to /qa/.
V 09/07/21 (Tue) 09:30:15 No. 8467
>>8465 >>8466 There's an overlap in the content of
>>>/qa/ and
>>>/aut/ that I want to resolve:
If
>>>/aut/ is for discussion of things in the real world then doesn't the blogging on
>>>/qa/ belong there?
If
>>>/qa/ is for hobbiest discussion then shouldn't the tech, music and food threads be moved out of
>>>/aut/ ?
But then you have to consider that there's a strong overlap between blogging and discussion of hobbies/activities.
So perhaps the name Seasonal Blogging is inaccurate?
Then there's also the legacy name for
>>>/qa/ as questions and answers that doesn't describe it's purpose other than the meta history of a
>>>/qa/ board. Meta is mostly relegated to
>>>/b/ though.
In the past I titled the board "quality anime" this doesn't describe it well either and it may be better to just drop the acronym system of naming /qa/ to something that more literal such as
/qa/ - Hobbiest Discussion . Though if this were the route to take then surely we would need a better name.
But my general purpose of writing this is that I think that /qa/'s purpose could be made clearer by offloading some more random aspects and less focused aspects of
>>>/qa/ onto
>>>/aut/ in the future.
V 09/07/21 (Tue) 09:03:06 No. 8465
I'm thinking about a few things
Tech discussion and the music discussion on
>>>/sum/ should be moved into
>>>/qa/ >>>/qa/70728 Will end up being locked on the seasonal rotation date to be remade. Will go on whatever board the remaker(who is not me) chooses. I personally support >>>/aut/ being used for this. At the end of Higurashi Sotsu if there's no new WTC anime then
>>>/cry/ should be locked and hidden so that the activity is focused onto
>>>/jp/ or
>>>/qa/ . There doesn't need to be an entire board devoted to Ciconia given kissu's size.
Anonymous 09/01/21 (Wed) 21:06:05 No. 8203
>>8202 The only two that I think I could disagree(or not even disagree but reword) with are 1 and 5.
1 because a site doesn't need to have a purpose, it doesn't have to be better or doesn't even need to be good. It can suppose a purpose to others or merely market itself as the true alternative to X. Hence why I think that the free speech imageboards are popular. More what I'm trying to say is that I would reword 1 to say
An imageboard .
5 because moderation and features can compensate for the fragmentation problem. An overboard removes the problem of fragmentation and a mod team who focuses on organizing and instructing others about the right way to do thing. Boards are simply an organizational tool, not a way to separate communities. This of course depends on how you view image-boards as either belonging to the 4/8chan model(each community is seperate) or how I think kissu works, each board is used by the same group and they can selectively hide whatever content they don't want to see (removing /sum/ for having too much real life content or /ec/ for being too sexual).
Anonymous 08/28/21 (Sat) 05:59:57 No. 8176
>>8172 >You probably have the footer-form focused This reminded me, very rarely I'll refresh a page -- typically after opening up my browser -- only for the page to seemingly fail to load, all but the footer. This fixes itself by refreshing the page again, but I figure I might as well mention it happening, at least on my end.
I might be misremembering, but I'm pretty sure I've had this happen on both my PC and phone - waterfox and Firefox mobile.
Anonymous 08/27/21 (Fri) 21:33:15 No. 8172
>>8170 OK, I think I actually understand what might be it. Not browser related.
You probably have the footer-form focused which by design prevents the QR from being opened.
Anonymous 08/11/21 (Wed) 18:18:46 No. 8029
Doing some other stuff at the moment so my explanation is limited, but it's been my perception for the past week+ that there are some threads scattered around that are out of place. Related to my thread >>>/poll/1554 3 weeks ago.
>>>/b/3588 >>>/qa/57322 >>>/qa/65289 >>>/qa/70728 >>>/qa/62191 They'd probably do better to be focused more and placed into the season rotational boards. However, I can't really call them problematic as much as they are distorting the primary function of /b/ and /qa/. Removing them at a time when they're being used might be more disruptive than productive.
The /b/ one though, I'm pretty sure that one has to be moved. /b/ isn't a blogging board.
Feedback #6: Culture, Optimizations and Improvements Anonymous 08/11/21 (Wed) 00:23:19 No. 8005 [Reply]
This is the next iteration of Kissu's feedback discussion AKA devblog. Though there's not much in the way of dev there are still bugs to fix. This iteration will hopefully focus on admin-blogging or tool development.
Important Note: If you want a new software feature (or a really any sort of change) provide me with a detailed proposal. It must be at least 3 sentences long and tell me why it's needed.
Updates: Software Nerd Things [/det]
Administration [/det]
Future Tasks: Software Post too long. Click here to view the full text.
Anonymous 07/08/21 (Thu) 04:50:32 No. 7612
I made a few major changes to how things work. I've abandoned or postponned the idea of implementing infinite scroll in favor of these changes.
Anonymous 06/01/21 (Tue) 23:07:31 No. 7361
i'll be uploading changes to the caret behaviour changing where the cursor lands after interacting and a bit of logic to prevent the QR from opening if you're focusing on the bottom form from No.xyz presses.
Anonymous 05/29/21 (Sat) 21:35:09 No. 7303
>>7300 ctrl+enter:
If no submission forms exist will do nothing
If focusing on a footer form element and QR is open will focus on the footer form
If focusing on a QR form element and footer form exists will focus on the QR form
If Footer-Form and QR-Form exists and not focusing on anything it will prioritize QR form
or something like this... will submit the form you've selected or if nothing is selected will prioritize QR
Anonymous 04/25/21 (Sun) 21:09:06 No. 6959
I can't see the post you made since verm's tarded and forgot to put /trans/ back on before deleting, but Andy's /pol/shit was completely unrelated to the thread and deserved to be nuked. When talking about UFOs sure it's selfish to expect people to not talk about conspiracies related to foreign countries in their theories. At the same time it's not selfish to expect people to not act like "FUCK THIS SHIT MAN YOU GOTTA FOCUS ON THE REAL ISSUES CHECK OUT MY REAL HARD POLITICS ". I mean that's just expecting common decency.
Anonymous 04/22/21 (Thu) 08:27:15 No. 6901
>>6898 >>6900 One proof goes like this: Draw two complex planes, in one of which we'll be plotting inputs to the polynomial, and in the other we'll be plotting the outputs. If you move your point in the input plane around a sufficiently big circle, it can be shown that the point in the point in the output plane goes around the origin n times, where n is the degree of the polynomial. The number of times a loop goes around a point is called its winding number.
If you split the loop in the input plane into two smaller loops, you get two smaller loops in the output plane. If the dividing line between the two output subloops passes through the origin, that means you've found an input that gives an output of zero. Then you can use polynomial long division to find an (n-1)-degree equation with the remaining solutions. But usually there will not be a solution precisely on the dividing line. In that case, the winding numbers of both subloops should add up to the original winding number. If you further divide the loops into smaller and smaller pieces, focusing on the subloops that have nonzero winding number, you find smaller and smaller input loops whose corresponding output loops make smaller and smaller loops around the origin. As the loops become smaller, they approach points in the input plane which map to the origin in the output plane. These are your solutions.
Of course, making this all rigorous, especially the idea of winding numbers, is a bit of work.
Anonymous 03/30/21 (Tue) 18:20:12 No. 6665
>>6652 There's way too much to talk about here, but I'll focus on the more pressing matter: what's the point of forcing anonymous when everyone knows each others' nicknames? What's the overlap between these people and those who tout the virtues of anonymity?
Anonymous 03/27/21 (Sat) 07:06:05 No. 6597
Altered the way highlights are handled. Might break something I haven't considered.
V 03/24/21 (Wed) 03:04:23 No. 6565
>>6563 because vichan wasn't made for an overboard. This is a forgotten feature on the new ui since I designed it specifically to work with an overboard. I should add this in as the 78th bug to fix. If you use vichan ui then it's tough luck..
>>6562 It's
noko sage so it takes out the noko and just leaves sage.
>>6561 Even if it doesn't directly tackle the root issue, removing things around it helps deal with it and coerces someone who might ad-hom to focus on substance.
>>6557 I like to see lots of different colors on the page and if people use noko, sage, email and default then it leaves a pretty rainbow of purple, green, blue and yellow across the screen.
Makes the posters contribute to the design of the site in a more impactful way. But in terms of actual usage, without a lot of people using noko in the first place it puts a lot of emphasis on the few that do. Maybe in a long distant future I'll readd it...
V 03/24/21 (Wed) 02:42:01 No. 6556
No this won't make any one person less identifiable, but it will make noko not a thing that people can target for promoting pointless tribalism, but focus on meaningful differences in culture.
Anonymous 03/19/21 (Fri) 14:27:43 No. 6499
>>6498 >>6497 This is fucking stupid and I'd like to know the full reasoning behind such asinine idea. First he wants people to lurk and learn and whatever the fuck else he wants with his ineffectual new UI but the precise microsecond someone makes a mistake that can be easily avoided A WHOLE FUCKING FEATURE IS REMOVED FROM THE SOFTWARE.
MAKE THE STATIC REPLY BOX DIFFERENT BY HAVING ALL THE OPTIONS
ADD A DETAILED CREATE THREAD FORM TO EVERY BOARD
STOP BULLSHITTING ABOUT MODERNIZING UI
YOU ARE A BAD WEBSITE DESIGNER
IT IS BECOMING EVER CLEARER
YOU WANT TO FORCE YOUR SHIT RICING ON EVERYONE
THE UI CONTINUES TO HAVE STUPID PROBLEMS
YOU ARE A BAD WEBSITE DESIGNER
FIX THIS SHIT UI ONCE AND FOR ALL
YOU ARE A BAD WEBSITE DESIGNER
YOU ARE A BAD WEBSITE DESIGNER
Post too long. Click here to view the full text. Anonymous 03/19/21 (Fri) 10:00:25 No. 6492
There is still no title option for replies and you should focus on fixing that glaring oversight first.
Anonymous 03/17/21 (Wed) 23:46:52 No. 6445
Instead of chasing around design issues I'm building up a list of things myself and others have problems with in the design. At the end of March I'll correct this list of issues. If problems still exist I'll make another list. By doing this I'm hoping changes will have more focus and be less erratic... more likely to resolve problems then create new ones.
>>6441 >Grotesque content(realistic or fictional, text or images) will not be allowed on any boards. This includes but is not limited to: guro, methods of suicide or plans to harm others Anonymous 01/22/21 (Fri) 04:16:11 No. 6128
Issues to resolve:- [Reply] shows up in thread on click - Entering threads will sometimes not focus on targeted post after navigation - Post Form width on mobile transition - Board layout settings in sidebar - Thread to have article/index settings - Lower post form images
Anonymous 01/06/21 (Wed) 21:44:24 No. 6062
This will be added with a list of issues to solve.
Bugs and Technical Issues 01- Clarify and fix errors when cookies don't match up with UI type 02- image previews/post previews on site navigation 03- Navigating from home into thread shouldn't keep hover open 04- options and name field not clearing on empty post submit 06- Options position 07- Post chain settings not being set, don't remove jump feature 21- Thread titles not matching between navigation and server requests 13- don't tab into filname input when no file active 14- Dragging text inside of the QR --
20- Sidebar closed menu with catalog closed 18- Remove highlight in catalog, reduce OP highlights Post too long. Click here to view the full text. Anonymous 01/05/21 (Tue) 20:00:34 No. 6051
>>6050 Now what needs to change:
Luna style isn't so bad but Hazuki style is the only one that's relaxing. Kissu style feels the wrong color compared to original.kissu.moe. I suppose I might be biased, as purple is among my favorite colors, and this purple is quite pleasing.
We will start with the left frame.
Again, you are wasting space. There are more things in it even though it feels like nothing was added. I had to compare side by side to realize what was added here that didn't allow you to move that disgusting bottom-right menu to where it clearly belongs.
And now there are variations depending on whether one is in the home page, in a board or in a thread.
...... I am going to focus in the most bloated, which is when one is browsing a board. Note however, an issue with the left frame @ home page. Pic related.
The simplest fix that will earn you a couple of precious pictures:
LINE UP THE BOARD NAME WITH THE FEED AND SIDE BAR TOGGLE.
I assume you don't because some board's names are too big and that offers a slight leeway in writing the boards in that space.
IT IS A MATTER OF VISUAL AESTHETICS.
Post too long. Click here to view the full text. Anonymous 01/04/21 (Mon) 23:41:16 No. 6002
>>6001 it boils down to the UI being haphazard and messy very unfocused/all over the place
the design is strait out of the 90's for better or for worse
Anonymous 01/02/21 (Sat) 18:22:50 No. 5940
Last items
- Bug on overboard vs boards - Custom CSS Storage bug - Previews being effected by catalog.css - mobilenav bug(.numbers) - mobile preview issue - hover box on non-existent items errors - catalog focus wider and offset to left:10px - Home previews multiple scrollbars appearing - outline:none; on threads - TypeError: Cannot create property 'mobile' on string 'Thread Props 404 - json 404 - returnReactAsStringObj err - Build Composite err - TemplaterBridge err - thread err on /jp/thread/12037' - TemplaterBridge err - thread double err on /jp/thread/12037 - deleted no box - delete shouldn't delete delete ?
- dependancy updates Post too long. Click here to view the full text. Anonymous 12/11/20 (Fri) 10:04:32 No. 5753
Going to publish a Ruby(for fun) tool to fix kissu's cite table. I don't think anyone will need it for their vichan servers, but I haven't publicly released anything in a while and haven't touched Ruby before.
Also will fix the truncation issue in
>>5354 if viewed through the 2.0-UI. This will wrap up all bugs I can touch at this time with my current setup.
>>5749 We'll see next month, but I tend to like to focus everything in this thread since it's easier for me to notice issues if it's all in one place.
Anonymous 12/11/20 (Fri) 05:29:21 No. 5728
>>5727 Have you tried escaping into media from people? Maybe a bit of ignoring the rest of the world and focusing on entertainment is what you need.
Anonymous 11/30/20 (Mon) 11:11:55 No. 5365
1. - You are following all the phases of "malcontent image board admin syndrome" to the letter. First it's new imageboard software to make things
look fresh, now it's misguided attempts at attracting more users to make things [i]feel/i] fresh, the latter brought about from reminiscing about good old days. Makes me worry about this site's longevity.
2. - You had a chance to do what you're thinking of way earlier in the year when henrietta attempted moving here. You squandered it. Now what you'll get out of this measure is turn /jp/ very literally into the coping grounds of someone who has irritable bowel syndrome in real life and is chasing the dopamine rush and validation he got when he was banned on the 4 for "literal shitposting" over 6 years ago. What will happen to kissu is exactly what happened to ota, regardless of your "inb4" warning people you already know what analogy they'll be tracing: there will be a torrent of low-effort contentless threads with outdated shock value, not enough people to turn them into something interesting, the original userbase of the site exhausts itself trying to keep a balance between the awful shit that's made their site home, eventually the balance breaks, the original population becomes increasingly discontent and you'll get spam wars where whatever remains of the original userbase attempts to restore balance by keeping at least one thing bumped (in ota it was love live), until even that's not enough and they turn on you.
Look at ota now and what's there and tell me without reservations that encouraging an environment where trash threads are the norm is worth losing whatever niche you're carving for yourself just so you have the pleasure of stumbling upon the statistic abnormality of "trash thread turned good". Unless you undertake the even more exhausting and arbitrary task of actively culling and deleting those threads as they begin dominating the front page and establish quotas, which is even more ridiculous a concept, but hey it's never been tried so who knows if it might work. I predict you'll become a babysitter overnight, though.
3. - People who don't come to kissu won't come because they don't want to. If it's not because it's too serious, it'll be for some other reason. Stop browsing ota and realize the admin of that place is acting maliciously and has successfully killed both whatever remained of /jp/ pre-AoC as well as other attempts at a site like it and like this (the ne
Post too long. Click here to view the full text. Anonymous 10/14/20 (Wed) 22:38:36 No. 4783
>>4779 that's to make it easier to highlight and refocus on posts from the overboard in the future version of the thread pages and make it easier to work with scenarios that involve posts from multiple boards on one page, but it causes issues with vichan pages.
I suppose for the timebeing it can omit the boardname+dash
Anonymous 10/05/20 (Mon) 23:50:52 No. 4699
notifications on mobile(chrome specifically) have to be done using the webworker notification system.
https://notifications.spec it's based around persistence meaning even when an application is in the background you can still be notified for it. I believe this also allows for people to be notified of kissu even when it's not in a tab on moba and desktop.
This isn't in my design philosophy currently and was always just an extra on the roadmap. I'd like to keep it simple and focus on making everything work before adding and tuning something intrusive.
Verniy ## Admin 09/15/20 (Tue) 08:55:23 No. 4446
I have to start concerning myself with money.
I stalled it out for a long time but I'm paying an affordable rent now and my savings are going to start getting chunked out every month. Is it possible to make a poverty level living on imageboards alone? It's time to put it to the test... I'm pretty certain getting a full-time job will mean quitting imageboards or making it a background project. I've got a commitment towards this site and community that goes beyond myself and want to continue as long as I can afford it. I want to focus on this site as long as possible and keep living a life of flexible hours and see Kissu gain in quality.
I'd like to give some sort of reward to donors and am thinking that it might be ideal to privately give out a whitelist key that exempts you from any vpn/proxy/tor/phone/agent bans and captchas. There are some ways I could do it without a sort of cash down deposit, but this is by far the most practical and my current donors are very generous and I want to give them something special.
This is the donation page should you be interested
https://www.patreon.co Money is an important priority in this stage of my work here, but it's more important for me to finish off the last security feature and create the whitelist keygen. At this point Kissu is in a state of management, maintenance and bug fixes, where beta features are ironed out(new UI) and fixes/adjustments are made to Kissu's software. However, ultimately I'd like to get away from the software and start focusing on the brand and I'm not too far from this point where I have a software and financial model I want to push. One that doesn't let me compromise my beliefs and stays true to the easy going atmosphere.
You won't have to deal with this kind of stuff right away, but in the next few weeks I'll start making a push. Until then, there are still bugs to solve and features to test.
Anonymous 09/06/20 (Sun) 17:59:52 No. 4341
The UI for multiupload is poor and gives threads with image spam an unfair advantage. Also gives way to dilution of individual post focus and drives interest away from text towards images
Anonymous 07/19/20 (Sun) 19:51:10 No. 3970
Wait, do you mean /b/ or kissu itself? Kissu averages around 120ish posts a day, /b/ is quite a bit slower as it's not terribly busy due to its focus on site development and stuff
Anonymous 07/11/20 (Sat) 02:28:16 No. 3937
4chan's issue is that everyone who cares ends up leaving. As soon as a spinoff shows up that offers the same content but with more focused moderation and administration, everyone fucks off there, leaving the original board a shell of its former self.
Anonymous 06/14/20 (Sun) 14:33:08 No. 3778
You know, at a first glance I would've said >>>/qa/43273 was a bit of an overreaction and that the definition would need to be revised, but looking at /trans/ I see it was entirely justified and even moderate.
I think there's still the element of easiness to be worked out, but I can't think of a way to formalize it and find scenarios where it's present in politics because the argument I make against politics is that it has a fundamental tendency to lack easiness.
Either way, ggwp.
>>3777 This maneuver would work if and only if we manage to get everyone discussing politics to focus on meta instead. Which would be fine by me, but it'd be hard to achieve.
Anonymous 06/02/20 (Tue) 22:05:00 No. 3710
final functionality for /home. Right clicking, handling adblock for banners, minor formatting and additional friends.
https://demo.verniy.ca/ I'll revisit it when I begin working on stylesheets, but no point in focusing on it too much. It's fulfilled it's role of working out core functionality.
For now the goal is to make /post-gen be able to output a fully functioning post. This means both contents and behavior. Server side and client.
When this is done I have to handle insertion of posts into threads, thread insertion into the frame and manipulations such as hiding and organization... I believe when this aspect is done that the major things to do will be modifications to vichan(archive adjustments such that it's possible to treat them as individual boards[though not a priority ghost posting could be done]) and an API server that will in the longterm take over vichan's tasks as needed.
Anonymous 06/01/20 (Mon) 06:42:23 No. 3698
https://demo.verniy.ca/ https://demo.verniy.ca/post-gen?com=sampl Banners require ublock being disabled.
Some comments:
- Post generator still not finalized, but you can see that the html structure I'm going for uses definition lists, lists, flexbox and CSS grid. This causes the structure of posts to be altered somewhat. A download button and native reverse imagesearch will be placed at the bottom of images. It's the intention for the native script to handle refined actions such as hiding through the context menu. In firefox options are in the menu, for chrome it's a submenu.
- There are also some obscured fields that are moved off the page for crawler algorithms to identify segments better
- The page functionality missing are things that will be needed later anyways. Such as making the thread level previews focus on the linked post on hover, right click menus to allow people to pin previews to the window.
- Bugs that I haven't yet touched are mostly on mobile and it's associated theme/page variation shown in the picture.
- A not very well tested feature is the notification systems. I've tried to get the webworker notifications setup, but couldn't figure it out and I don't have a good testing environment for this yet without the API server. Desktop notifications for posts added to the site work though and redirect to the page, but you won't see them at this point in time.
>>3694 I have some time again to touch on this and see if I can find out what's wrong with the banner loading
Anonymous 05/21/20 (Thu) 08:11:55 No. 3572
>>3551 reading
>>3555 it makes me think that the style of treating posts as a tree(or list) of connected nodes is flawed to either requires the reader to be accustomed with the preexisting system(non intuitive UI) or approximate it with redundant data(exploitable behaviour).
Ideally the posts should be visually represented as a graph somehow. That way no information about the structure has to be assumed by the reader and the index doesn't become bloated by approximations.
I don't think this is possible to do clearly without thinking in 3D space Making use of Z indices and a completely dynamic container form that also manages to impart full information while scrolling. But I have no idea how to represent this.
>>3566 Delisted those boards, added score counting, fixed poll viewing from the overboard. I probably won't add these changes to the repo as yet.
I have an idea on how I want to take kissu's software and really focus the experience, reduce my work load and do away with some of the problems that 4chan inspired imageboards have. I think that this unsaid solution is only possible with an SPA which enables rapid page changes.
I'd rather put it into practice/draw it than explain and I have to finish off the homepages first.
I list 12 issues that need to be covered for the the navigation bars, homepage, errors, basic styles and navigation to be completed/easily expandable. Going by my work rate for the past few days that's a 2 or 3 day job but giving myself 5 until Monday. After which I'm going to add some 4taba features and begin working on golang to write a better API service.
Anonymous 05/14/20 (Thu) 01:18:19 No. 3532
>>3522 It runs against my design decisions of increasing accessibility. Double click, even if it's harder to implement, is more in line with what I'm aiming for. The software shouldn't discourage people for who they are or what their platform is. Removing links would be working against my design.
>>3528 Hover preview is an existing useful functionality working on only one type of link in the page(cite links). It can be expanded further and become a more uniform action that people think of when they see an underline. The (embed) system native to 4chanX is kind of not a good choice here because there's no plan as yet for inline comment nesting and I'd rather discourage huge threads to sway things away from the use of generals, at least until the site is too big.
In the topic of inline behaviour, I'm seeking to move away from this in general and the coming update focuses more on subwindow based expansion rather than inline. This is the design choice I'm trying to run with.
Anonymous 05/08/20 (Fri) 09:59:14 No. 3433
Working on drafts for generating the custom error page first. SSR and client side hybrid. This means the basic 404 header, sidebar, topbar and banners.
Anonymous 05/03/20 (Sun) 21:56:39 No. 3228
The problem I have with keeping it on the homepage is diluted function and lack of focus. If I cater it to being a makeshift overboard then the landing page doesn't show off the site very well. 8kun's homepage for example tells me nothing about the site and is a cluttered mess. I want to avoid this
Anonymous 05/01/20 (Fri) 09:33:39 No. 3128
I don't know how many sequence diagrams I've done, but there's probably only 5-7 ones left. I need to check over my list again.
Anonymous 04/20/20 (Mon) 01:46:54 No. 2932
For the eventual and probable mod UI I'm going to think about completely dropping the idea of a mod mode browsing UI and focus more on the selection of items from lists and searches.
Anonymous 04/17/20 (Fri) 17:56:56 No. 2900
The idea of focusing more on users rather than admins also suggests that many new features of the site will come through the UI rather than the server. So it needs to be built in a really modular way. Each segment needs to be a removable chunk of information
Anonymous 04/17/20 (Fri) 17:49:27 No. 2899
I'm throwing all of the old systems into the trash and using it as reference. Stylesheets may change a bunch too. Likely will be something like that, but more expandable and modular. The old design promotes a rigid view of what's allowed and not. Needs to focus on users rather than admins.
Anonymous 04/17/20 (Fri) 04:28:32 No. 2895
The reason I'm focusing so much on the software right now is because two things need to be done for kissu to expand. Better front end behavior and a second main board to focus /qa/'s content.
Anonymous 04/16/20 (Thu) 21:55:34 No. 2889
Meaning that I don't intend for Kissu to compete with 4taba. Adding new features to vichan feels like progress, but it's actually just bogging the site down with barely maintainable crap. kotatsu is more specialized in focus than general purpose vichan so have to do more work in refining kissu to the minimum needed feature set.
Anonymous 04/10/20 (Fri) 21:48:59 No. 2802
Different use case than expanding media that shrinks the amount of possibilities a user. Using that aspect of the UI is to focus your attention on images and view as an image catalog. I don't feel like I gain from this unless I'm reading a book.
Anonymous 04/10/20 (Fri) 06:57:39 No. 2798
>>2794 Not so sure about how I feel about that kind of image expansion for non-video things. It is pretty nice for webms and such, but seems like it'd get annoying for regular images/gifs. Although it'd especially be annoying for anything that's audio focused that I'd just want to expand and see in the background.
Anonymous 04/06/20 (Mon) 06:57:49 No. 2759
>>2758 Either that or I annoyed everyone into submission with rants. Even then I don't know where they'd go.
Probably 4chan if I think about it which is all the more reason to have something unique about the site. Front end Software isn't at a state yet where I can spend my dev hours advertising anyways. Back-end doesn't matter too much until bottleknecks are hit. When kissu has a responsive user experience and good visual aesthetic the user cap should be raised.
http://s11.flagcounter.com/more30/CexS/
from the flag counter on the site I feel like we hit a plateau of advertising effort to result. Part of it is visual feel, other parts is what I'd complain to be a too much of a documentation heavy culture on the site and not discussion focused. Much of /qa/ is paranoid and not 'tistic.
Anonymous 04/06/20 (Mon) 06:31:34 No. 2757
yeah, I stopped promoting board for the past 2 week to focus on software
Anonymous 04/04/20 (Sat) 00:02:44 No. 2719
>>2717 Reading over this again, I think I'm getting too caught up in the flaws with vichan and not appreciating what's good about it.
I think
>>2711 is a more professional point of view on how to take it. I know that most big websites are written with multiple languages to suit multiple goals. It means I have to do less wheel reinventing as well.
So I'm caught between:
A) Working with a small NodeJS imageboard and expanding it into something more serious.
B) Expanding the PHP, NodeJS and NGINX solution that will be put into place as a short term fix into something more serious. Despite me sounding more favorable to A before it's probably more practical to do B.
B will be done for the next period of time anyways.
Though I think this JSON focused imageboard idea is interesting, it means that I have to throw away a lot of vichan's pretty not terrible code. I'm more familiar with PHP than NodeJS at this point anyways.
>>2718 Post too long. Click here to view the full text. Anonymous 04/03/20 (Fri) 20:39:34 No. 2717
Deleted a bunch of posts to refocus my thoughts
>>2712 I thought the idea behind lynx was to take away vichan's thunder, copying whatever would make it easy to move to. They certainly advertised on a bunch of sites running it. I'm not sure if Lynx is the right pick if I would change though. I described to someone in Steam that if Vichan is McDonalds then Lynx would be BurgerKing.
I'm leaning towards a lightweight framework written by someone who just wanted to make an imageboard.
>>2711 I think it could be done. It will end up being a lot of nginx rules for each PHP page that is an exception. It corssed my mind. It feels really hacky but that's the objective here. Never know. The posts.php page and mod.php pages are the two big things needed.
Just will take another API for post form info to retrieve: ```<input type="hidden" name="hash" value="35c4710c8f598a2
b5e8c468270f909af7cceb
030">```
I wanted to use 4taba's kotatsu as a potential kissu2.0 but he chose the wrong tools for the job.
Ilia's Hikkaba could be of promise
https://github.com/magicxor Post too long. Click here to view the full text. Anonymous 04/02/20 (Thu) 18:45:45 No. 2702
i think server has a lot of junk on it that could be removed for that, but it's not at the point of worry. In post time, by comparison it's still faster than sites that are bigger. Kissu runs on cheap and the expectation is everyone has a faster computer.
Anonymous 03/19/20 (Thu) 20:56:27 No. 2454
These changes cut the size
{"no":24101, Post too long. Click here to view the full text. Anonymous 03/19/20 (Thu) 20:28:43 No. 2453
each thread is this much information:
http://json.parser.online.fr/
{"no":24101, Post too long. Click here to view the full text. Anonymous 03/13/20 (Fri) 11:03:21 No. 2380
Decided to rangeban the frog guy through NGINX on the banners since he outlived his usefulness as a tester.
I believe the banners project is in a point where I can turn away from it and use it as a tool. There are some other features that can be added like banner voting and perceptual hashing.
Most of the software related things to do now are dry concepts like making the software installable, size issues on polling boards, fixing the file-board reply counter bug and removing some poor python-php integrations i did.
On the side of considering how to best rework vichan, it's trying to get some monetary value out of things.
not sure what to do with 4taba, but I think i'll let it idle until there's another Artanis patch out.
Also need to make a blog so I can write these things ramblings down somewhere more easily ignorable for people who don't care. Maybe put a few view/click based ads on it.
However, I'm trending more towards a front-end rework that fixes visually unsettling aspects in both the HTML and yotsuba, dark and kissu themes.
I'm thinking that even there all the frontend could be done using a multipage React application and would look cleaner, run faster(so much bloat JS, and require the server to do less keeping growth costs down
Post too long. Click here to view the full text. Anonymous 02/10/20 (Mon) 23:47:40 No. 2243
>>2241 It's just a suggestion. Few people are writting the foundational aspects of websites nowadays and leave it in the hands of the people who are more interested in creating foundations than applications. Laravel and Symfony are examples of these. The benefit is that your project gets to be less focused on the site itself meaning spend more time adding features. Also readability and it's easier for people to contribute if they want
oyasumi-chyuu.fun Anonymous 02/09/20 (Sun) 01:13:56 No. 2234 [Reply]
This is the thread for discussing improvements, fixes, additions and development of oyasumi-chyuu, a site made to house /qa/'s collection of OPs and EDs. Up until now I've been working on it on my own, but it was always intended to become a place that the community can develop together. I haven't asked for any help on it because I wanted to learn how to do things on my own, but now that it has most of the main features that I wanted to add its a good time to open it up. If it seems like a very basic site, thats because it is. I'll make a github soon enough so everyone can look at the actual files and help improve them as needed, but if theres something that you think needs to change and you know how to do it, by all means please help.scraper . Its been suggested that we make a scraper that will pull information about a given OP from a site like anidb.net. The scraper will put the information into the post form for the user, so all they really have to do it select the file to upload, select which OP it is and hit Go. This will both minimize the chance for errors and wrong or inaccurate information entered, and will make uploading a series of videos from the same show much simpler. Aside from this, its also been suggested to make a scraper that will collect more information about the song used in each OP itself )such as the song name, artist) and display that. There are a couple other sites that have information on the complete soundtrack from different shows so we can pull from there. Plus, if we pull from those sites, it gives us a more accurate number of how many OPs and EDs a series has so we can collect them all. I haven't looked too deeply into this, so if theres anybody out there with any experience or who knows how to do this your help would be greatly appreciated!